Live Data
In Data Refinery Workbench, Live Data refers to customer data that is either remediated (clean) or available for remediation. Customer data can be imported directly as Live Data, or can be imported as Workflow data, which eventually becomes Live Data after processing through the Workflow.
Table of contents
Defining a Data Object
A Data Object is the representation of an individual Live Data record. Data Objects contain Data Attributes of the record, which adhere to the structure defined by the object’s associated Entity Type. A Data Object represents one and only one data record identified by its key Data Attribute values. These Attribute Type keys are specified in the Data Object’s Entity Type definition. Importing data records multiple times results in one Data Object, provided that the key values are unchanged. Data Objects also have a unique Data Object ID. Importing data records using the ID of existing Data Objects allows for updates to key values.
Data Objects reside in the Live Data store, where they can be retrieved through Live Data Search by specifying the Entity Type and attribute values. See the Live Data Search page to learn more about searching for Live Data.
How to Create a Data Object
Live Data is created through the import of data records as Data Objects, or as Workflows upon the completion of the Workflow process. Data Object and Workflow creation can be accomplished through the Workbench API. See the Data Refinery Workbench API Reference for more information.
How to Update a Data Object
Data Objects can be updated in two ways: through Workflows created from existing Data Objects, or through direct import of data records with updated values.
Data Objects are updated through Workflows where editing is done on a copy of the record data. Workflows can be created directly through import of data records. When importing a Workflow with literal record data, key attributes of the record are used to search for an existing Data Object. If a Data Object is found, the unique Data Object ID is associated with the new Workflow and the Workflow Attributes are populated with the import record data. If a Data Object is not found by the time the Workflow End Transition is applied, a new Data Object is created for that updated record.
When a Workflow is created, the Data Object attributes are copied to the new Workflow as Workflow Attributes. Users with access to the Workflow Data, based on the Workflow status and User Group membership, modify these Workflow Attributes. When a Workflow End Transition is applied (e.g. after review and approval), the updated Workflow Attribute data is copied back into the Data Object.
Multiple active Workflows can exist for the same Data Object. This supports scenarios where different team members are focused on different data record fields. However, care should be taken to coordinate multiple Workflows since changes made by completing one Workflow can be overwritten by unchanged data in a subsequent Workflow.
Data Objects can also be updated via direct import. Importing the same data record multiple times results in the same Data Object, based on the data record key attribute values. Hence, a Data Object can be updated by updating non-key data record field values for subsequent import. Moreover, the import data file format includes an optional column for Data Object IDs. Importing updated data records to existing Data Objects using Data Object IDs provides a way to update key record attributes through direct import.
Data manipulation for updating Data Objects through direct import would occur outside of Data Refinery Workbench.